Association of GSTP1, COMT, and TPMT with testicular cancer (TC) treatment outcome after cisplatin-based chemotherapy (CBCT).

Abstract

Listen

Translate article

344 Background: Genetic variation in GSTP1, COMT, and TPMT, which are genes important to the metabolism of cisplatin, may affect TC treatment outcome. We hypothesized that single nucleotide polymorphisms (SNPs) of these genes causing increased cisplatin metabolism may be associated with worse treatment outcome. Methods: TC patients who had CBCT, ≥ 1 year of follow-up since initiation of CBCT, and genotyping completed as part of a genome-wide association study were included. After excluding markers with genotyping call rates of < 90%, we analyzed 3, 43, and 44 SNPs each for GSTP1, COMT, and TPMT respectively. We evaluated three endpoints: refractory disease (RD) as defined by radiographic or serologic evidence of TC after initial CBCT; progression-free survival (PFS); and overall survival (OS). We studied SNP association with these endpoints assuming additive, recessive, and dominant genetic models and adjusted p-values for false discovery rate to account for multiple comparisons. Results: 137 patients (82.5% white, median age of 31 years) were recruited. Among them, 81.7% had non-seminoma and 68.6%, 8.0% and 22.6% were in good, intermediate, and poor IGCCC prognostic group respectively. 33.6%, 62.0%, 0.7%, and 2.9% had EP, BEP, TIP, and VIP respectively as initial CBCT with a mean total cisplatin dose of 345.3 mg/m2. After initial CBCT, 8.0% developed RD; among the remaining 92.0%, 23.8% had recurrent TC after a mean follow-up of 46.8 months. We found no statistically significant associations of our SNPs with PFS or OS. However, the COMT rs2073743 G allele conferred a higher risk of RF. It was present in 40.9% of patients with RD versus 21.1% of those without. Assuming recessive inheritance, patients with rs2073743 GG had over 20-fold increased odds (odds ratio = 22.69, (95% CI 3.3 – 155.8) ) of RD compared to those who carry at least one C allele. After adjusting for race, IGCCC, primary site of disease, type of CBCT, and total cisplatin dose, this association was no longer statistically significant. Conclusions: COMT rs2073743 may be associated with RD, but replication and validation studies will be imperative to confirm this finding.